News & Views | Published:

Cell biology

Recycling in sight

Nature volume 501, pages 4042 (05 September 2013) | Download Citation

Vision requires the continuous recycling of photobleached pigments. An atypical form of a degradative pathway called autophagy seems to participate in this process in retinal pigment epithelial cells.

Vision begins with the absorption of photons by light-sensitive photoreceptor cells in the retina. When photons arrive, chromophore molecules in the photoreceptors undergo conformational changes and trigger the phototransduction cascade, which converts light into electrical impulses that travel to the brain and are transformed into the images that we see. To sustain vision, the chromophore 11-cis-retinal must return to its original conformation through a process known as the visual cycle. Writing in Cell, Kim et al.1 demonstrate that one process that contributes to the proper functioning of the visual cycle is an atypical form of autophagy — the 'self-eating' pathway through which cells recycle their components by degrading them in cellular organelles called lysosomes.

The visual cycle involves tight regulation of the interaction between photoreceptors and the adjacent retinal pigment epithelium (RPE) cell layer; the latter nourishes photoreceptors and improves the quality of the optical system of the eye by absorbing scattered light. Another key role of the RPE is degradation and recycling of the photoreceptor outer segments (POSs), which are damaged by the impact of millions of photons every day. Each morning, the distal 10% of POSs are engulfed (phagocytosed) by the RPE and degraded inside lysosomes. Whereas this phagocytic process has been known for decades, the intimate molecular and cellular mechanisms of chromophore and photoreceptor recycling have remained unclear.

Kim et al. demonstrate that selected proteins implicated in autophagy are necessary for the degradation of POSs. Specifically, they show that POS phagocytosis coincides with an increase in the levels of the autophagy marker protein LC3-II, and that it involves an atypical autophagy pathway called LC3-associated phagocytosis (LAP). Moreover, in the RPE, LAP requires typical autophagy regulators such as the Atg5 protein (Fig. 1). However, it is independent of the proteins of the autophagy pre-initiation complex that in typical autophagy mediate the formation of autophagosomal vesicles to sequester cytoplasmic components2, consistent with previous work3. Given that, in LAP, the phagosomal vesicles carrying the material to be degraded have already formed, it is perhaps not surprising that this process is independent of the autophagy pre-initiation complex.

Figure 1: Role of autophagy in vision.
Figure 1

Light enters the eye and strikes the retina, where it activates rhodopsin pigments in the membranes of photoreceptor outer segments (POSs). This induces the release of both the bleached pigment (opsin) and a molecule of all-trans-retinal, which is converted to vitamin A (all-trans-retinol) before forming the chromophore 11-cis-retinal in the adjacent retinal pigment epithelium (RPE) cells through the visual cycle. The binding of 11-cis-retinal to opsin restores the photoactivable pigment rhodopsin, which can react again with light and start the phototransduction cascade. Light also damages POSs, which are subsequently phagocytosed and eliminated by RPE cells. Kim et al.1 report that, after phagosome formation, an atypical autophagy pathway involving the proteins Atg5 and LC3 triggers phagosome fusion with the lysosome, resulting in POS degradation. The degradation products exit the lysosome and are recycled to sustain the visual cycle. Blood is another source of vitamin A.

Maturation of phagosomes usually involves their acidification and the acquisition of degradative enzymes to break down the ingested material. Why, then, are autophagy proteins required for LAP? It is thought that recruitment of autophagy proteins to the phagosomal membrane allows its rapid fusion with lysosomes, enhancing degradation4. In support of this view, Kim and co-workers observed no defects in phagocytic uptake in Atg5-deficient RPE cells. However, the phagosomes in these cells failed to migrate towards the basal side of the cells, where fusion with lysosomes occurs. Furthermore, maturation of the protease enzymes required for degradation was impaired, and the expression of lysosomal-membrane glycoproteins in these phagosomes was reduced, suggesting impaired phagosome–lysosome fusion. At least in this setting, therefore, LAP does seem to promote trafficking of phagosomes and their fusion with lysosomes, ensuring rapid recycling of POSs.

The visual cycle is a transcellular process by which RPE cells maintain the supply of chromophores for the regeneration of visual pigment in photoreceptors (Fig. 1). When photoreceptors absorb light, 11-cis-retinal is converted to all-trans-retinal through isomerization. All-trans-retinal is then released from the membrane-bound receptor opsin and reduced to all-trans-retinol (vitamin A), which diffuses to the intercellular space and enters the adjacent RPE cells, where it is transformed back into 11-cis-retinal5. Other sources of vitamin A are the blood and POS phagocytosis by the RPE6.

Kim and co-authors show that autophagy proteins mediate 11-cis-retinal recycling following POS phagocytosis. Although many aspects of this process remain unclear, chromophore recovery from discarded POSs may represent a highly efficient mechanism for recycling. Each day during POS shedding, around 10% of the total ocular pool of retinoids passes through the phagolysosomal system of RPE cells. The present paper suggests that at least some of the 11-cis-retinal can be recovered and used to regenerate visual pigment without the involvement of intercellular transport systems or enzymes of the visual cycle, which are often mutated in retinal diseases6.

The authors further show that mice specifically lacking Atg5 in their RPE have impaired vision and reduced chromophore levels. Interestingly, previous work7 has found that retinal ageing is associated with decreased autophagic activity and a parallel reduction in night vision; these traits were mimicked by Atg5 deletion in retinal precursors. Also, in the lens, defects in autophagy-protein function result in age-related cataracts8. And retinal ganglion cells lacking Atg5 and Atg4, another autophagy-associated protein, show increased sensitivity to optic-nerve damage9.

A common feature of these autophagy-impaired mutations is the accumulation of toxic products, suggesting a defect in intracellular quality control. For instance, accumulation of lipofuscin, a cellular waste product derived from the incomplete digestion of POSs, is frequently observed in human retinal diseases such as age-dependent macular degeneration6. This disease is caused by a primary malfunction of the RPE, which leads to photoreceptor death and subsequent blindness.

Kim et al. did not investigate whether there was an increase in lipofuscin levels in their Atg5-deficient mice. But photoreceptor numbers remained normal in these mutants, even after 7 months. And the authors could reverse the effects of Atg5 deletion on vision by giving the animals retinoid supplements. These results suggest that downregulation of autophagy primarily causes alterations in the visual cycle and not defects in intracellular quality control, as was previously thought.

Vitamin A and retinoid supplements have also been used to treat some retinal disorders associated with visual-cycle defects and ageing6. Intriguingly, retinoid derivatives increase the activity of chaperone-mediated autophagy10 — a form of autophagy that is upregulated in the retina as a compensatory response to age-associated decreases in macroautophagy, one of the more common autophagic pathways7. Regardless of the details, two points are becoming clear: autophagy proteins play a part in maintaining vision fitness, and retinoids show potential for the treatment of retinal diseases.


  1. 1.

    et al. Cell 154, 365–376 (2013).

  2. 2.

    , & Nature Cell Biol. 15, 713–720 (2013).

  3. 3.

    , & Nature Rev. Mol. Cell Biol. 13, 7–12 (2011).

  4. 4.

    et al. Nature 450, 1253–1257 (2007).

  5. 5.

    Chem. Rev. 101, 1881–1896 (2001).

  6. 6.

    , , & Annu. Rev. Pharmacol. Toxicol. 47, 469–512 (2007).

  7. 7.

    et al. Aging Cell 12, 478–488 (2013).

  8. 8.

    et al. J. Biol. Chem. 288, 11436–11447 (2013).

  9. 9.

    , , , & Cell Death Differ. 19, 162–169 (2012).

  10. 10.

    et al. Nature Chem. Biol. 9, 374–382 (2013).

Download references

Author information


  1. Patricia Boya is in the Department of Cellular and Molecular Biology, Centro de Investigaciones Biológicas, Consejo Superior de Investigaciones Científicas, 28040 Madrid, Spain.

    • Patricia Boya
  2. Patrice Codogno is in INSERM U845, Necker Growth and Signaling Research Center, Université Paris Descartes, 75014 Paris, France.

    • Patrice Codogno


  1. Search for Patricia Boya in:

  2. Search for Patrice Codogno in:

Corresponding author

Correspondence to Patricia Boya.

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing